A Bayesian approach for noisy matrix completion: optimal rate under general sampling distribution

DOI10.1214/15-EJS1020zbMath1317.62050arXiv1408.5820OpenAlexW3098122928MaRDI QIDQ2340879

Pierre Alquier, The Tien Mai

Publication date: 21 April 2015

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1408.5820

zbMATH Keywords

Bayesian analysismatrix completionGibbs sampleroracle inequalitylow-rank matrixPAC-Bayesian bounds

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Generalized linear models (logistic models) (62J12) Random matrices (probabilistic aspects) (60B20) Learning and adaptive systems in artificial intelligence (68T05) Random matrices (algebraic aspects) (15B52)

Related Items

Tight risk bound for high dimensional time series completion ⋮ 1-bit matrix completion: PAC-Bayesian analysis of a variational approximation ⋮ An oracle inequality for quasi-Bayesian nonnegative matrix factorization ⋮ A reduced-rank approach to predicting multiple binary responses through machine learning ⋮ Bayesian uncertainty quantification for low-rank matrix completion ⋮ User-friendly Introduction to PAC-Bayes Bounds ⋮ Matrix completion with nonconvex regularization: spectral operators and scalable algorithms ⋮ Concentration of tempered posteriors and of their variational approximations ⋮ An efficient adaptive MCMC algorithm for pseudo-Bayesian quantum tomography ⋮ Bayesian matrix completion approach to causal inference with panel data ⋮ Pseudo-Bayesian quantum tomography with rank-adaptation ⋮ Matrix factorization for multivariate time series analysis ⋮ Numerical comparisons between Bayesian and frequentist low-rank matrix completion: estimation accuracy and uncertainty quantification ⋮ On the exponentially weighted aggregate with the Laplace prior ⋮ Oracle posterior contraction rates under hierarchical priors ⋮ Sharp oracle inequalities for low-complexity priors ⋮ Estimation bounds and sharp oracle inequalities of regularized procedures with Lipschitz loss functions ⋮ On the properties of variational approximations of Gibbs posteriors

Uses Software

• tmvtnorm
• Jellyfish

Cites Work

• Unnamed Item
• Unnamed Item
• Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion
• Statistical learning theory and stochastic optimization. Ecole d'Eté de Probabilitiés de Saint-Flour XXXI -- 2001.
• PAC-Bayesian bounds for sparse regression estimation with exponential weights
• Some PAC-Bayesian theorems
• Parallel stochastic gradient algorithms for large-scale matrix completion
• Noisy low-rank matrix completion with general sampling distribution
• Exact matrix completion via convex optimization
• Bayesian Methods for Low-Rank Matrix Estimation: Short Survey and Theoretical Study
• The Power of Convex Relaxation: Near-Optimal Matrix Completion
• Restricted strong convexity and weighted matrix completion: Optimal bounds with noise
• Sparse single-index model